Data-Driven Motion Mappings Improve Transparency in Teleoperation

Abstract

A teleoperation system with high transparency enables the operator to focus on completing the task at hand instead of on controlling the robot. We previously proposed that modifying the mapping from human movement to desired robot movement might improve the transparency of teleoperators in ways similar to adding sensory feedback. Specifically, we created non-Cartesian motion mappings that correct for systematic reaching errors made by humans, so that the robot motion resembles the operator's intent rather than his or her produced movement. This article presents a study that compares subjects' performance on a virtual teleoperated targeting task under three different motion mappings: a Cartesian-scaling motion mapping that is typically implemented in teleoperators, a corrective variable-similarity motion mapping that is fit to aggregate data from subjects in a previous study, and a corrective variable-similarity motion mapping that is fit to calibration data collected from each subject. Twelve participants reached toward 120 targets under each of the three motion mappings with balanced random presentation order and a washout task between conditions. Subjects were able to complete the targeting task with higher accuracy in the initial direction of robot motion, at higher speeds, and with more natural and efficient reaching movements under the variable-similarity motion mappings. Subjects also overwhelmingly preferred the variable-similarity motion mappings. These results indicate that subjects experienced a higher level of transparency when using the virtual teleoperator with the variable-similarity motion mappings than with the standard Cartesian mapping. Therefore, mappings that correct for systematic errors in human motion, such as the variable-similarity motion mappings tested here, should be considered in teleoperator design.

Data-driven body–machine interface for the accurate control of drones

“…Indeed, it is known that humans make directional errors when relying only on proprioception to estimate the spatial location of their limbs, and that these errors are proportional to the distance to the body centerline ( 53 , 54 ). Building on this knowledge, previous studies showed that nonlinear transformations of the users’ arm movements led to faster and more precise control of a robotic arm than a simple scaling ( 14 , 55 ). Further studies will be needed to understand the role of more complex mappings to extend the results of this work.…”

“…Successful teleoperation requires robust and reliable control interfaces. A well-defined interaction should be transparent ( 14 , 15 ), rely on intuitive command inputs to ensure rapid proficiency and minimize the task-associated workload ( 16 ), and provide appropriate feedback (visual, auditory, haptic) to strengthen the awareness of the operator ( 17 ). A number of existing interfaces already allow interactions with robotic devices.…”

Personalized Telerobotics by Fast Machine Learning of Body-Machine Interfaces

“…During this phase, the user has no control over the robot trajectory, and they are asked to move in an instinctive way, as if they were controlling the flight direction with their body. This approach is similar to the ones seen in [30], and [31]. During this phase, the human pilot is sitting on a stool.…”

“…Recent works investigated the natural interaction strategy of humans controlling drones through behavioral experiments [28], [29], to determine common mapping strategies, yet with a supervisory control paradigm. User-specific mapping functions have been proposed for anthropomorphic manipulators [16], [30] and ground robots [31]. Nonetheless, the first case is limited to homologous systems, while the second does not investigate the effect of such a method on learning and usability.…”

Hand-worn Haptic Interface for Drone Teleoperation

“…The human-robot mapping, as it is implemented at the moment, is a simple scaling of the hand position. It could be improved to data-driven mappings, as in [43] to help the users to learn it faster. Moreover, to allow the deployment of our system in environments in which a MoCap is not available, the use of wearable sensors to track hand motion should be investigated.…”